WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-users

RE: [Xen-users] pv_ops 2.6.31.6

Hi,

I have recompiled the kernel and the performance is much better so thanks for 
that.

What does that setting do anyway?

Thanks

Ian 


-----Original Message-----
From: Pasi Kärkkäinen [mailto:pasik@xxxxxx] 
Sent: 23 January 2010 19:15
To: Ian Tobin
Cc: Olivier B.; xen-users@xxxxxxxxxxxxxxxxxxx
Subject: Re: [Xen-users] pv_ops 2.6.31.6

On Sat, Jan 23, 2010 at 07:12:12PM -0000, Ian Tobin wrote:
> Good point!
> 
> I think i have overlooked that but i do recall seeing it in the instructions. 
>  Would that slow it down then?
> 

Yeah, at least on my 32b systems it'll make dom0 crash in half an hour or so.. 

CONFIG_HIGHPTE=n has been stable for me.

-- Pasi

> 
> Ian
> 
> 
> -----Original Message-----
> From: Pasi Kärkkäinen [mailto:pasik@xxxxxx] 
> Sent: 23 January 2010 18:32
> To: Ian Tobin
> Cc: Olivier B.; xen-users@xxxxxxxxxxxxxxxxxxx
> Subject: Re: [Xen-users] pv_ops 2.6.31.6
> 
> On Sat, Jan 23, 2010 at 06:18:05PM -0000, Ian Tobin wrote:
> > Thanks, ill do the test and wait for RAM before doing anything further.  No 
> > errors in dmesg and its a 32bit Dom0.
> >
> 
> Do you have CONFIG_HIGHPTE=n in .config? 
> 
> If it's enabled (=y) then your dom0 kernel will definitely crash.. 
>  
> -- Pasi
> 
> > Ian
> > 
> > 
> > 
> > -----Original Message-----
> > From: Pasi Kärkkäinen [mailto:pasik@xxxxxx] 
> > Sent: 23 January 2010 11:56
> > To: Ian Tobin
> > Cc: Olivier B.; xen-users@xxxxxxxxxxxxxxxxxxx
> > Subject: Re: [Xen-users] pv_ops 2.6.31.6
> > 
> > On Sat, Jan 23, 2010 at 09:38:28AM -0000, Ian Tobin wrote:
> > > Hi,
> > > 
> > > Appologies for the delay.
> > > 
> > > Xm top shows that a DomU (windows 2008) was using 100% cpu however it 
> > > wasn't actually doing anything 
> > > but i killed it off so no DomUs were running and the system was still 
> > > painfully slow until its rebooted but it still slows down.
> > >
> > 
> > So you're saying the system was slow even after the windows 2008 guest was 
> > killed? Something definitely wrong there..
> > Any errors in "xm dmesg" ? or in dom0 "dmesg" ?
> > 
> > Was this 32bit dom0?
> > 
> > > 
> > > I have discovered the RAM is not the recommended RAM kit for this 
> > > motherboard so i have ordered 8gb 
> > > of the correct RAM type.  Do you think this would be the cause?
> > > 
> > 
> > Who knows.. maybe. 
> > 
> > Make sure you run memtest86+ for a long time to make sure the RAM is ok.
> > Sometimes the memory errors will show up only after a couple of _days_ of 
> > running memtest.
> > 
> > http://memtest.org/download/4.00/memtest86+-4.00.iso.zip
> > 
> > -- Pasi
> > 
> > > thanks
> > > 
> > > Ian
> > > -----Original Message-----
> > > From: Pasi Kärkkäinen [mailto:pasik@xxxxxx] 
> > > Sent: 22 January 2010 11:23
> > > To: Ian Tobin
> > > Cc: Olivier B.; xen-users@xxxxxxxxxxxxxxxxxxx
> > > Subject: Re: [Xen-users] pv_ops 2.6.31.6
> > > 
> > > On Fri, Jan 22, 2010 at 10:52:04AM -0000, Ian Tobin wrote:
> > > > Ive got a major performance issue.  If i boot up the dom0 without 
> > > > starting xen and copy a large file to it via ftp its fine.
> > > > 
> > > > As soon as i run xend start the Dom0 gets slower and slower until the 
> > > > server is at a crawl so its unusable.
> > > > 
> > > > Anyone else getting slow response?
> > > > 
> > > 
> > > I haven't seen that.
> > > 
> > > Please monitor your dom0 with "top" and also with "xm top". 
> > > 
> > > What do they reveal? Does some process start eating more and more CPU 
> > > time? 
> > > Does some process leak memory? Does dom0 have iowait? 
> > > 
> > > Did you limit dom0_mem=512M or similar? 
> > > 
> > > -- Pasi
> > > 
> > > > 
> > > > 
> > > > 
> > > > -----Original Message-----
> > > > From: xen-users-bounces@xxxxxxxxxxxxxxxxxxx 
> > > > [mailto:xen-users-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of Olivier B.
> > > > Sent: 22 January 2010 08:34
> > > > To: xen-users@xxxxxxxxxxxxxxxxxxx
> > > > Subject: Re: [Xen-users] pv_ops 2.6.31.6
> > > > 
> > > > For me restore/live migration doesn't work.
> > > > 
> > > > For dom0 I use the version 2.6.31.6 00751-g600545, with vanilla 
> > > > 2.6.31.12 pv_ops version on domU, and debian xen 3.4.2-2
> > > > 
> > > > On restore I obtain that :
> > > > 
> > > > [49532.764004] <4>------------[ cut here ]------------
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] <4>WARNING: at arch/x86/xen/time.c:180 
> > > > xen_sched_clock+0x7c/0xaf()
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] <d>Modules linked in:
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  nf_conntrack_ipv4
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  nf_defrag_ipv4
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  xt_state
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  nf_conntrack
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  dm_snapshot
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [last unloaded: scsi_wait_scan]
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] Pid: 30422, comm: kstop/0 Tainted: G        W  
> > > > 2.6.31.12-dae-xen #1
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] Call Trace:
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8104e0d1>] warn_slowpath_common+0x88/0xb6
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8104e121>] warn_slowpath_null+0x22/0x38
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8100dc4f>] xen_sched_clock+0x7c/0xaf
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8101829d>] sched_clock+0x9/0xd
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8106b73b>] sched_clock_cpu+0xa7/0x168
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff81047d4b>] update_rq_clock+0x26/0x48
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff81048c11>] try_to_wake_up+0xac/0x2af
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8101135d>] ? retint_restore_args+0x5/0x6
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff81048e34>] default_wake_function+0x20/0x36
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8103bcdd>] __wake_up_common+0x58/0xa2
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8106120e>] ? wq_barrier_func+0x0/0x36
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8103e07d>] complete+0x49/0x73
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8106122e>] wq_barrier_func+0x20/0x36
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff81060d8d>] worker_thread+0x156/0x20d
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8100dd1f>] ? xen_restore_fl_direct_end+0x0/0x1
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff81065b7f>] ? autoremove_wake_function+0x0/0x5a
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff81060c37>] ? worker_thread+0x0/0x20d
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff81065759>] kthread+0x9b/0xa3
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff810119ea>] child_rip+0xa/0x20
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff81010bac>] ? int_ret_from_sys_call+0x7/0x1b
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff8101135d>] ? retint_restore_args+0x5/0x6
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004]  [<ffffffff810119e0>] ? child_rip+0x0/0x20
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] <4>---[ end trace bbe4ba0e56e4a4ae ]---
> > > > [49532.764004] BUG: recent printk recursion!
> > > > [49532.764004] <4>------------[ cut here ]------------
> > > > 
> > > > 
> > > > But except that, it works fine.
> > > > 
> > > > Olivier
> > > > 
> > > > Pasi Kärkkäinen a écrit :
> > > > > On Thu, Jan 21, 2010 at 09:01:21PM -0000, Ian Tobin wrote:
> > > > >   
> > > > >>    Hi,
> > > > >>
> > > > >>
> > > > >>
> > > > >>    Quick question, is the 2.6.31.6 kernel in Jermeys tree stable 
> > > > >> enough for a
> > > > >>    live environment?  Ive been playing with it and seems to work 
> > > > >> quite well.
> > > > >>
> > > > >>     
> > > > >
> > > > > It works for many people.. so please keep using it, and report any 
> > > > > issues/bugs found. 
> > > > >
> > > > > Also make sure you monitor the changelogs, it's still under 
> > > > > development so you want to
> > > > > upgrade every now and then to get the latest bits.
> > > > >
> > > > > xen/master branch changelog:
> > > > > http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git;a=shortlog;h=xen/master
> > > > >
> > > > > Whole git tree changelog:
> > > > > http://git.kernel.org/?p=linux/kernel/git/jeremy/xen.git
> > > > >
> > > > > -- Pasi
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > Xen-users mailing list
> > > > > Xen-users@xxxxxxxxxxxxxxxxxxx
> > > > > http://lists.xensource.com/xen-users
> > > > >   
> > > > 
> > > > _______________________________________________
> > > > Xen-users mailing list
> > > > Xen-users@xxxxxxxxxxxxxxxxxxx
> > > > http://lists.xensource.com/xen-users
> > > > 
> > > > 
> > > > 
> > > > _______________________________________________
> > > > Xen-users mailing list
> > > > Xen-users@xxxxxxxxxxxxxxxxxxx
> > > > http://lists.xensource.com/xen-users
> > > 
> > > 
> > 
> > 
> 
> 



_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users

<Prev in Thread] Current Thread [Next in Thread>